contrasting effects of long term versus short-term
TRANSCRIPT
Contrasting effects of long term versus short-term nitrogen addition on photosynthesis and respiration in the Arctic
Martine J. van de Weg, Gaius R. Shaver, Verity G. Salmon
This article was accepted for publication in Plant Ecology.
van de Weg, M.J., Shaver, G.R., and Salmon, V.G. 2013. Contrasting effects of long term versus short-term nitrogen addition on photosynthesis and respiration in the Arctic. Plant ecology. 214(10): 1273-1286. Available from DOI:http://dx.doi.org/10.1007/s11258-013-0250-6
The final publication is available at Springer viahttp://dx.doi.org/10.1007/s11258-013-0250-6
1
1
Title: Contrasting effects of long term vs. short-term nitrogen addition on photosynthesis and 1
respiration in the Arctic. 2
3
Authors: Martine J. van de Weg1, 2, Gaius R. Shaver2 and Verity G. Salmon3 4
5
Author addresses: 6
1. Amsterdam Global Change Institute, Vrije Universiteit Amsterdam, De Boelenlaan7
1085, 1081 HV Amsterdam.8
2. The Ecosystem Centre, Marine Biological Laboratory, 7 MBL Street, Woods Hole,9
Massachusetts, USA.10
3. Department of Biology, University of Florida, Gainesville, Florida, USA.11
12
Corresponding author details: 13
Martine J. van de Weg 14
E-mail: [email protected] 15
Tel: 0031-205986877 16
17
18
Word count (minus abstract and references): 5341 19
Word count abstract: 247 20
21
2
2
22 Abstract 23
We examined the effects of short (<1 to 4 years) and long-term (22 years) nitrogen (N) and/or 24
phosphorus (P) addition on the foliar CO2 exchange parameters of the arctic species Betula 25
nana and Eriophorum vaginatum in northern Alaska. Measured variables included: the 26
carboxylation efficiency of Rubisco (Vcmax), electron transport capacity (Jmax), dark 27
respiration (Rd), chlorophyll a and b content (Chl), and total foliar N (N). For both B. nana 28
and E. vaginatum, foliar N increased by 20-50% as a consequence of 1 to 22 years of 29
fertilisation, respectively, and for B. nana foliar N increase was consistent throughout the 30
whole canopy. However, despite this large increase in foliar N, no significant changes in 31
Vcmax and Jmax were observed. In contrast, Rd was significantly higher (>25%) in both species 32
after 22 years of N addition, but not in the shorter-term treatments. Surprisingly, Chl only 33
increased in both species the first year of fertilisation (i.e. the first season of nutrients 34
applied), but not in the longer-term treatments. These results imply that: 1) Under current 35
(low) N availability, these Arctic species either already optimize their photosynthetic capacity 36
per leaf area, or are limited by other nutrients; 2) Observed increases in Arctic NEE and GPP 37
with increased nutrient availability are caused by structural changes like increased leaf area 38
index, rather than increased foliar photosynthetic capacity and 3) Short-term effects (1-4 39
years) of nutrient addition cannot always be extrapolated to a larger time scale, which 40
emphasizes the importance of long-term ecological experiments. 41
42
43
Keywords: nitrogen use efficiency, fertilisation, LTER, Alaska, chlorophyll, canopy, leaf 44
mass per area 45
46
47
3
3
48 Introduction 49
The productivity of Arctic tundra ecosystems is limited by cold temperatures, short 50
growing seasons and low nutrient (nitrogen (N) and phosphorus (P)) supply (Shaver and 51
Chapin 1980, 1986; Chapin et al. 1995). Although the N deposition rates in the Arctic are 52
relatively low compared to industrialized and temperate regions in the Northern hemisphere, 53
N deposition has increased with the global rise in anthropogenic N emissions the past century 54
(Bobbink et al. , 2010). Furthermore, warming of the Arctic is expected to increase N 55
availability, as warming experiments in Alaska have shown increased available N though 56
increased mineralization rates, or through permafrost thawing (e.g. Johnson et al. 2000; 57
Shaver et al. 2001; Keuper et al. 2012). Additionally, studies on Arctic watersheds already 58
have observed higher export rates of different N forms (i.e. nitrate, ammonium, dissolved 59
organic nitrogen) most likely as a consequence of warming of tundra and increased thawing 60
of permafrost (Frey et al. 2007; McClelland et al. 2007). Given the anticipated environmental 61
change, N availability in the Arctic is expected to increase in the coming century. 62
63
On a plot stand scale, it has been well established that long-term N (together with P) 64
addition in Arctic tundra increases the biomass, leaf area index (LAI), gross ecosystem 65
production and ecosystem respiration (Shaver et al. 1998; Boelman et al. 2003; Mack et al. 66
2004). Furthermore, long-term N and P addition causes a shift in species composition, with 67
an increase in shrub cover, while bryophytes and forbs are reduced (Shaver and Chapin, 1991 68
1995; Bret-Harte et al. 2002; Hobbie et al. 2002 Zamin and Grogan, 2012). Less is known, 69
however, about the long-term effects of N and P addition on dark respiration (Rd) and net 70
photosynthesis (Anet) at the leaf level. Moreover, the effects of increased N on the more 71
fundamental determinants of photosynthetic capacity, the maximum carboxylation velocity 72
(Vcmax) of the enzyme ribulose-1,5-bisphosphate carboxylase/oxygenase (Rubisco) and the 73
4
4
maximum rate of electron transport (Jmax), remain minimally explored for tundra vegetation. 74
Maximum Anet, Vcmax, Jmax and Rd are tightly related to N content, mainly because of the high 75
investment of N in the C3 photosynthetic apparatus (Field and Mooney, 1986; Evans, 1989), 76
and high involvement of N-rich proteins in maintenance respiration for protein turnover 77
(Penning De Vries 1975). Previous N and/or P addition experiments in Arctic tundra (lasting 78
1-4 years) showed a diverse array of effects on (maximum) Anet and Rd for different species. 79
For example, Matthes-Sears et al. (1988) found an increase of total foliar N after four years of 80
nutrient addition with N and P, but this had no consequence for Anet in Betula nana and Salix 81
pulchra. This is similar to a recent finding of Heskel et al. (2012) who found no effect of four 82
years of N and P addition on maximum Anet for B. nana and Eriophorum vaginatum, but 83
increased levels of Rd in the Alaskan tundra. Contrastingly, Oberbauer et al. (1989) found 84
higher rates of maximum Anet after one and two years of NPK fertilizer in B. nana and Ledum 85
palustre (but not in Carex biggelowii), while Chapin and Shaver (1996) similarly found 86
higher foliar N and maximum Anet values in B. nana and E. vaginatum after four years of N 87
and P addition, but not in L. palustre and Vaccinium vites-ideae. 88
89
Overall, the effects of N and/or P addition to tundra vegetation on foliar gas exchange 90
parameters are not straightforward. Furthermore, field nutrient addition experiments that 91
exceed 5 or 10 years of nutrient addition are rare, not only in Arctic tundra. It is questionable 92
whether results regarding maximum Anet and Rd, or the proportional investment of N in the 93
photosynthetic apparatus, can extrapolated from a relatively small number of years to a 94
prolonged period of elevated nutrient supply (> 20 years). Whether the photosynthetic 95
nitrogen use efficiency (PNUE) changes after long periods of increased N supply is of 96
particular interest, since in increasingly more vegetation models or up-scaling exercises the 97
parameters Vcmax and Jmax, as well as Rd, are scaled by the amount of foliar N (Friend et al. 98
5
5
2009; Zaehle and Dalmonech 2011). Therefore, if the PNUE or Rd-N relationships change 99
with changes in N availability or N, this might have consequences for the accuracy of 100
predictions of future carbon uptake. Finally, most of the nutrient addition studies mentioned 101
above only included fully sunlit leaves, from the top of the tundra canopy. In general, foliar N 102
on an area basis (Narea), as well as Vcmax- and Jmax on an area basis decline with decreasing 103
light throughout a canopy, though the pattern of the foliar traits throughout the canopy does 104
not follow the patterns of decrease in irradiance in 1:1 proportion (e.g. Meir et al. 2002; 105
Niinemets 2007). Whether the investment of foliar N in the photosynthetic apparatus 106
throughout the canopy in the Arctic differs under fertilized and an unfertilized condition has 107
received little attention. 108
109
In this study we investigated the effects of different durations and rates of N and/or P 110
fertilizer addition on the foliar CO2 exchange parameters of the two common Arctic tundra 111
species B. nana and E. vaginatum. More specifically, the aims of this study were: 1) to 112
investigate the effects of short and long term N addition on the foliar CO2 exchange 113
parameters Vcmax, Jmax, and Rd, as well as foliar N, the foliar chlorophyll content (Chl) and 114
leaf mass per area (LMA); 2) to investigate whether the relative N investment in Vcmax, Jmax, 115
Rd and chlorophyll changes with different nutrient addition amount and different durations, 116
and 3) to investigate whether the relative N investment in these CO2 exchange parameters 117
differs at different canopy positions in fertilized and unfertilized tundra. 118
119
Methods 120
Research area and species 121
For this study we sampled two N and P addition experiments that are located in moist 122
acidic tundra within the Arctic Long Term Ecological Research (LTER) site in the northern 123
6
6
foothills of the Brooks Range, Alaska (68°38'N, 49°43'W, elevation 720 m). In both 124
experiments, nutrients are added to plots of 50 m2, as NH4NO3 and P as P2O5 every spring 125
following snow melt. Site 1 was installed in 1988, and details can be found in Bret-Harte et 126
al. (2001). From Site 1 the control (CT), N addition, P addition and N and P addition (NP) 127
treatments were used, with all treatments replicated in four blocks (Table 1). Site 2 was 128
installed in 2007, approximately 150 m south of Site 1, and includes a range of quantities of 129
N+P addition (up to 10 g N m-2 yr-1), with 50 m-2 plots which are replicated in four blocks as 130
well. From Site 2 the treatments that receive respectively 5 and 10 g N m-2 yr-1 were selected 131
for this study, together with the accompanying control treatment (Table 1). Additionally, we 132
added a very short term N+P fertilisation treatment to Site 2. For this we installed four extra 1 133
m2 plots to which 10 g m-2 N and 5 g P m-2 was added on 30 May 2010 (Table 1). 134
The two species included in the study are the dwarf shrub Betula nana L. and the 135
sedge Eriophorum vaginatum L., which are common throughout the whole Arctic region 136
(Britton, 1966). These two species were chosen because they were both abundant enough in 137
the control and nutrient addition plots from both experimental sites to sample for this study, 138
as in particular many evergreen species have decreased in (relative) abundance in the NP 139
plots of Site 1 (Gough et al. , 2012). 140
141
Experimental design 142
Gas exchange measurements were conducted on fully sun-lit leaves that were 143
collected from all of the research plots in Table 1 between 5 and 15 July 2010 (i.e. around the 144
peak of the growing season). Per species, treatment, and plot, two A-Ci curves (measuring 145
maximal photosynthesis rates at a range of intercellular CO2) were performed, leading to 8 146
replicates per treatment and 64 measurements in total for each species. For the E. vaginatum, 147
6-10 leaves were used and measurements took place between 8-15 July 2010 for site 1 and 148
7
7
between 5-15 July 2010 for site 2. For B. nana, 3-4 mature leaves attached to the twig were 149
used and the measurements for site 1 took place between 6-8 July 2010, and between 5-6 July 150
2010 for site 2. Furthermore, for B. nana, the respiration rate of the corresponding twig part 151
that had been in the cuvette was measured afterwards, in order to correct the values from gas 152
exchange measurements. In addition, for both species the gas exchange measurements were 153
spread in a way that CT and nutrient addition samples were alternated throughout the day and 154
per photosynthesis system (see below or description of the gas exchange measurements). 155
Although this resulted in a not-complete random design, it avoided one of the treatments 156
being measured in a cluster in time or per photosynthesis system. 157
Between 10 and 30 July in 2011, we additionally measured foliar gas exchange on B. 158
nana leaves that were growing at different light regimes (i.e. the top, middle and bottom of 159
the canopy) from a CT and NP plot at Site 1 (Table 1). The light regime of the three canopy 160
positions were characterized using three quantum sensors (LI-90 Li-Cor Inc, Lincoln, USA) 161
that were placed at different canopy positions in both the CT and NP. Diurnal relative 162
photosynthetic active radiation (PAR) and average PAR was calculated using the average 163
photon flux density logged every 10 minutes (CR1000X, Campbell Scientific Ltd, Logan, 164
UT, USA) throughout the month long measurement period (Fig. 1). We sampled 6 replicates 165
of twigs with 3-4 leaves positioned around each of the quantum sensors (i.e. these leaves 166
were experiencing the same light regime) for gas exchange measurements, leading to 36 167
samples in total. 168
169
Gas exchange measurements 170
For the measurements in 2010, three open portable photosynthesis system (Li-Cor 6400, Li-171
Cor Inc, Lincoln, USA), fitted with LED light sources (6400-02B Red/Blue Light Source, Li-172
Cor, Inc, Lincoln, USA), were used for the A-Ci curves, following the procedural guidelines 173
8
8
in Long and Bernaccchi (2003). The CO2 concentrations inside the chamber ranged from 50 174
to 2000 ppm, and leaf temperature was set at 20 °C (average Tleaf was 20.1 °C ± 0.05, n= 175
128). Gas exchange measurements were conducted only between 10:00-18:00, to avoid any 176
diurnal artefacts on leaf functioning. In the field, B. nana branches and E. vaginatum leaves 177
were detached and immediately re-cut under water in the field, in order to reconstitute the 178
water column. The sample was subsequently brought to the Toolik Field Station lab (at ~ 1 179
km distance) in order to start the measurements. This method, as opposed to conducting the 180
A-Ci curves in the field on attached leaves and twigs, limited trampling of the vegetation in 181
the long-term nutrient addition plots and it avoided taking the sensitive photosynthesis 182
systems out in unfavourable weather. Most importantly, conducting the gas exchange 183
measurements in the field station lab enabled us to do this at nearly the same temperatures 184
(20 °C), since the Li-Cor 6400 can control the leaf chamber temperature only within a limited 185
range from ambient temperatures. Tests on non-detached and attached branches and leaves 186
showed the shape or values of the A-Ci curves stayed the same before and after detachment. 187
Following the A-Ci curves, Rd was measured, after keeping the leaf in darkness for a 188
minimum of 20 minutes to avoid transient changes in CO2 release associated with post-189
illumination changes in metabolism (Azcón- Bieto and Osmond 1983). Gas exchange 190
measurements made in 2011 followed the same procedures though the Li-Cor 6400 was fitted 191
with a lighted conifer chamber (6400-22L, Li-Cor, Inc, Lincoln, USA). 192
After the gas exchange measurements, leaf area was measured using a desktop 193
scanner and Winfolia software (Regent Instruments Inc, Canada). The leaves were then dried 194
to a constant weight at 60 °C and weighed. Subsequently, the leaves were ground and 195
analysed individually for total C and N content with a Perkin-Elmer Series II 2400 CHNS/O 196
Analyzer (LECO Corporation, U.S.A.). The leaf mass per area (LMA) was calculated by 197
9
9
combining the leaf area and leaf dry weight measurement and the LMA values were used to 198
convert mass-based leaf parameters area-based ones. 199
200
A-Ci response curve analysis 201
We used a curve fitting routine (Sharkey et al. 2007) to analyse the A-Ci curves to calculate 202
Vcmax and Jmax on a leaf area basis. The curve fitting is based on minimum least-squares was 203
used in “R” (R Development Core Team 2008). The fits were obtained using the Farquhar 204
biochemical model of leaf photosynthesis (Farquhar et al. 1980; von Caemmerer 2000). The 205
enzymatic kinetic constants were taken from Table 1 in Sharkey et al. (2007), while the 206
parameters for the curvefiting to 20 ºC were scaled using temperature dependencies provided 207
by Bernacchi et al. (2001, 2002, and 2003). 208
209
Chlorophyll analyses 210
Chlorophyll content of the B. nana leaves was determined using leaf level reflectance 211
measurements, as the leaf tissue from the gas exchange measurements was not enough to 212
measure both foliar N and Chl from. Therefore, directly after the gas exchange measurements 213
and before drying, leaf level reflectance for each wavelength between 350 and 1100 (R350-214
R1000) was measured with a field portable spectrometer (Unispec, PP Systems, Amherst MA, 215
USA) and its accompanying bifurcated fibre optic cable and leaf clip. A calibration curve for 216
leaf level reflectance and chlorophyll content (a and b) was made with a subset of 50 B. nana 217
leaves from the research site. From these leaves, chlorophyll was extracted with N,N-218
dimethylformamide (DMF) and determined photospectrometrically as described in Porra et 219
al. (1989), after they were freeze dried and stored temporarily at -80°C. The mSR705 index 220
((R750-R445 )/(R705-R445)) and chlorophyll content were then used to create a calibration curve 221
(P<0.0001, R2= 0.67). With this calibration curve we determined the chlorophyll content of 222
10
10
the B. nana leaves from the gas exchange measurements based on leaf level reflectance. For 223
the E. vaginatum samples, the leaves were not suitable for leaf level reflectance 224
measurements (i.e. the leaf area did not fit in the leaf clip). Therefore, we collected a 225
representative subset of sunlit leaves from each plot (n=4) and their chlorophyll content was 226
determined directly at the research station using a Tris/acetone solution for extraction agent, 227
as described by Sims and Gamon (2002). 228
229
Statistics 230
Statistical analyses were performed in R (R Development Core Team, 2008). There 231
were no interactions of the blocks with the treatments throughout the suite of measured 232
parameters, therefore, the replicates per treatment were grouped together. To test for 233
significant differences between the treatments and their controls, we performed ANOVA’s 234
with post-hoc Dunnett’s tests. This form of post-hoc test compares between the control 235
treatment and all other treatments, as these were the contrasts of interest in this study. 236
237
Results 238
Foliar nitrogen content 239
Nutrient addition effects on foliar N were not consistent through time or across 240
species. Addition of both N and N+P increased Narea in the leaves of shrub B. nana with 42% 241
after 4 and ~50% after >20 years of fertilisation (P<0.001 and P=0.002, respectively), while 242
no increase was observed after the first year of N+P addition (Fig. 2). However, for the B. 243
nana that had received only 4 years of nutrient addition, only the treatment that had received 244
the highest dosage of N+P for 4 years (F10) had a significantly higher Narea values (P<0.05), 245
while Narea was not significantly higher than the control in the treatment that received half the 246
dosage of fertiliser (F05). Additionally, Nmass of B.nana was 20% lower than the control in 247
11
11
the P only treatment (P<0.05) for this species. For the sedge E. vaginatum there was no effect 248
of N and/or P addition when expressed on an area basis. However, Nmass for this species was 249
6% higher (P<0.05) with N only and 15% higher with N+P after > 20 years ((P<0.01, Fig. 250
2). For E. vaginatum, no significant influence of the short-term fertilisation on Narea or Nmass 251
was found. 252
253
Foliar CO2 exchange parameters 254
Despite the increase in Narea as a result of N and P addition, there was no significant 255
increase in Vcmax in B. nana, for any length of nutrient addition (Fig. 3a and 3b). In contrast, 256
B. nana Vcmax in the P-only treatment was 41% lower than in the CT treatment (P<0.05). For 257
E. vaginatum, there was no significant difference observed in Vcmax among the different 258
treatments in Site 1 or 2. A similar pattern was found for Jmax; for both species there was no 259
significant influence of nutrient addition on this parameter in either of the sites (Fig. 3c and 260
3d). In contrast to the photosynthetic parameters, Rd was 50% higher for B. nana and 27% for 261
E. vaginatum in the N+P treatment in Site 1 (P<0.05), but not for the N-only treatment (Fig. 262
3e and 3f). Finally, no significant differences in Rd between treatments and the controls were 263
found for E. vaginatum or B. nana in Site 2. (P= 0.43 and P=0.27, respectively). 264
265
Chlorophyll and LMA 266
The chlorophyll content on an area basis (Chlarea) was significantly higher for both 267
species in the YR1 treatment compared with their controls (P<0.05). In contrast, leaves from 268
both species that had experienced nutrient addition for more than one season showed no 269
significant increases or decreases in Chlarea (Fig. 4a and 4b). Similarly, for LMA only the 270
YR1 treatment in E. vaginatum was significantly lower than the control treatment (P<0.05) 271
12
12
(Fig. 4d), but no differences in LMA were found in other treatments for either of the 272
investigated species. 273
274
Different canopy positions 275
Because of the lower LAI at the CT plot, PAR levels at the bottom of the canopy was 276
higher than at the bottom of the canopy in the N+P plot (Fig. 1), and it was not possible to 277
collect leaves in the CT treatment that had been grown at the same low light levels as in the 278
NP treatment). Therefore, ‘bottom canopy’ leaves from the CT treatment cannot be compared 279
with ‘bottom canopy’ leaves from the N+P treatment directly. Taking all canopy positions 280
together, however, leaves from the N+P treatment had a higher foliar N than those from the 281
control treatment both on a mass and area basis for the whole dataset (P<0.001 and P=0.02, 282
respectively, Fig. 5a and 5b Student’s t-test). For the N+P treatment, the fully sunlit leaves 283
had similar Nmass values to those growing at lower light levels in the canopy, but significantly 284
higher LMA values (0.001, Fig. 5f). In contrast, no significant differences in LMA were 285
found in the CT treatment for the leaves from the different light levels. Chlarea did not differ 286
significantly between the treatments or between canopy positions (data not shown), but when 287
expressed on a mass basis (Chlmass), the leaves from the NP treatment that received the least 288
PAR, had higher values than the top canopy leaves (P>0.039 Fig. 5e, Student’s t-test). Vcmax 289
did not differ between the N+P or CT treatment for top canopy leaves, and the relationship 290
between irradiance and Vcmax was similar for the CT and N+P treatment. The low-light leaves 291
in the NP treatment had significantly lower Vcmax values than the top canopy leaves 292
(P=0.008, Fig. 5c). 293
294
Discussion 295
13
13
This study shows that leaf level photosynthetic capacity of B. nana and E. vaginatum 296
is insensitive to increases in leaf level N in the Arctic tundra. This suggests that some Arctic 297
species already maximize their photosynthetic capacity per leaf area under ambient nutrient 298
availability, or that they get limited by other nutrients when foliar N increases. Consequently, 299
the relationship between N and Vcmax or Jmax is not linear for the two common Arctic species 300
studied here. 301
302
Long term nutrient addition and foliar N 303
With both a relatively short (4 years) and long duration of the N and P addition, Narea 304
and Nmass increased in B. nana, and for E. vaginatum only on a mass basis. This is consistent 305
with other short-term (1-3 year) Arctic nutrient addition studies that showed increases of 306
foliar N (on either a mass or area basis) after N addition (Oberbauer et al. 1989; Chapin and 307
Shaver 1996; Shaver et al. 2001; Heskel et al. 2012), though some did not find this trend (e.g. 308
Van Heerwaarden et al. 2003). One difficulty with comparing nutrient addition studies is that 309
some studies only include fully sun-lit leaves, while in others the Narea and Nmass represent an 310
average of leaves from the whole canopy. The LAI and the presence of B. nana increase 311
substantially in moist acidic tussock tundra after several years of N (and) P addition (Shaver 312
et al. 2001; Street et al. 2007) and consequently, the bottom leaves in the canopy receive less 313
PAR (Fig. 1). These lower PAR levels cause the lowest leaves in the canopy to have a lower 314
LMA (Ellsworth and Reich 1993; Evans and Poorter 2001; Niinemets 2007; and Fig. 6d this 315
study), and can consequently decrease the bottom canopy Narea values, although Narea was not 316
significantly lower than the top leaves in the N+P treatment in our observations (Fig. 5b). 317
Nevertheless, if leaves of the whole N+P canopy are included in an average Narea value, the 318
leaves from the bottom of the canopy (with lower Narea values) could skew canopy averages 319
of Narea to lower average numbers because they contain proportionally less canopy leaves than 320
14
14
the average of a CT canopy. Indeed, for Alaskan tundra shrub communities, the amount of 321
total N in a canopy does not increase linearly, but asymptotic with increasing LAI (van Wijk 322
et al. 2005), which would result in lower average foliar N values at high LAI. Therefore, 323
comparing only the top canopy leaves (which will have received the same PAR regime) 324
between two treatments can be expected to show differences in foliar nutrients more obvious 325
then when canopy averages are compared. 326
327
Foliar N and CO2 exchange parameters 328
Firstly, given the generally conservative ratio between Vcmax and Jmax (Wullschleger 329
1993) it is not unexpected that both photosynthetic parameters show similar patterns with 330
Narea. It is more surprising that for both B. nana and E vaginatum, no increase in Vcmax or Jmax 331
was observed in the leaves with higher foliar N values from the N addition plots (Fig. 3a-3d). 332
In contrast, when foliar N is lower than under control conditions (in the P only treatment, Fig. 333
2), the photosynthetic parameters Vcmax and Jmax are also lower (Fig. 3a and 3c). In other 334
words, the relationship with N and photosynthetic parameters holds when N is decreased 335
from ambient conditions, but when it increases the relationship becomes non-linear until it 336
de-couples. In addition, the patterns of Vcmax and Jmax throughout the N+P and CT canopy 337
overlap in an almost continuous pattern (Fig. 5c and 5d ), even though the leaves in the N+P 338
treatment that received the least radiation had high Narea and Nmass values compared with the 339
leaves in the CT treatment (Fig. 5a and 5b). This shows how radiation levels are an important 340
determinant for the patterns of photosynthetic capacity throughout a canopy (Meir et al. 2002 341
2001; Niinemets, 2007), since even when foliar N is high, this N is not used for the 342
photosynthetic capacity when the average received levels of PAR are low. 343
Similarly, for both species no increase in Chl (Fig. 4a and 4b) was observed with Narea 344
increase, except for the first year of when the N+P addition. This first-year increase in Chl 345
15
15
did not coincide with an increase in Vcmax or Jmax, which could be a consequence of the first 346
year N and P addition (i.e. a transient short term effect where in the absence of previous 347
nutrient addition extra N first is invested in Chl). Like Vcmax and Jmax , the pattern of Chl 348
throughout the canopy (Fig. 5e) was overlapping between the CT and N+P treatment, and the 349
increased Chl in the bottom canopy N+P leaves can be attributed to lower irradiance levels, 350
rather than higher foliar N (Evans and Poorter 2001, Niinemets 2003). 351
The de-coupling of foliar with photosynthetic capacity contrasts with the studies 2-3 352
year N-addition studies from Oberbauer et al. (1989) and Chapin and Shaver (1996) who 353
found higher levels of maximum Anet after fertilisation in B. nana. However, both these 354
studies did not include measurements of the photosynthetic parameters Vcmax an Jmax, so there 355
is a chance their increase in maximum Anet is a consequence of changes in the stomatal 356
behaviour for example. Furthermore, de-coupling of foliar N and Vcmax has also been 357
observed with long-term (9 years) NPK addition in a nutrient poor bog in Canada and after 15 358
years of nutrient addition to a temperate forest (Bauer et al. 2004), which make our results not 359
unlike those from other ecosystems. The decrease in PNUE suggest that under ambient, low- 360
nutrient conditions of the Arctic, the N investment in foliar C-uptake is already optimal on a 361
leaf level scale, perhaps because these species (B. nana and E. vaginatum) have evolved 362
under low N availability. Alternatively, scarcity of other nutrients such as Mg2+, could be 363
limiting the photosynthetic capacity in the leaves with high foliar N. For example, Manter et 364
al. (2005) observed an increase in Rubisco (the enzyme involved in the first major step of 365
carbon fixation) in fertilised Pseudotsuga menziesii seedlings, but similar to our findings, the 366
activity of this Rubisco decreased with increasing foliar N. This Rubisco inactivation was 367
linked to a decreased relative availability of Mg2+, which led to Mn-induced Rubisco 368
deactivation. Additionally, Heskel et al. (2012) observed an increase in chloroplast area in E. 369
vaginatum and B. nana after N+P addition. Larger chloroplasts as a consequence of high N 370
16
16
supply has been correlated with decreased Rubisco specific activity and PNUE in other 371
species as well (Li et al. 2013), and is explained by a decreased ratio of mesophyll 372
conductance to Rubisco content and a lower Rubisco specific activity. It is likely that in our 373
study the increased foliar N lead to a similar pattern of increased Rubisco content with a 374
reduced activity, with no increase in Vcmax as a consequence. 375
It could be argued that the high dosages of N and P addition in our study (up to 10 g 376
m-2 yr-1 for N) do not resemble realistic magnitudes of increased N availability due to Arctic 377
warming. Indeed, Keuper et al. (2012), reported an increase of ~ 240 mg N m-2 in the rooting 378
zone of an Arctic bog following thawing permafrost, which is an order of magnitude lower 379
than our largest annual N application. Nonetheless, the decoupling of foliar N and 380
photosynthetic capacity itself is an important observation, since foliar N (or foliar N modelled 381
after N availability) is often used as a (linear) scalar for gross or net CO2 uptake (Thornton et 382
al. , 2007; Kattge et al. , 2009; Zaehle and Friend, 2010). If this decoupling happend at 383
relativley high foliar N values like in our study, we expect this decoupling to happen also at 384
more moderate increases of foliar N. Therefore, we think that not taking into account this N-385
photosynthesis de-coupling at increased N availability could lead to overestimations of the 386
photosynthetic parameters Vcmax and Jmax in CN-dynamic models. 387
As for Arctic stand-scale CO2 exchange, the de-coupled photosynthesis-N relationship 388
in the two Arctic species also implies that the observed increases in gross (and net) CO2 389
uptake on a ground area basis (such as measured with 1 m2 chambers) after N and/or P 390
addition in Arctic tundra are a consequence of increases in LAI, and not a consequence of 391
increased photosynthetic capacity per leaf area (Boelman et al. 2003, Street et al. 2007). 392
Indeed, 75 % of the variability in plot level CO2 uptake amongst Pan-Arctic vegetation types 393
could be explained by radiation levels and LAI alone, without having to consider foliar N 394
levels (Shaver et al. , 2007; 2013). In short, adding N (with P) increases ecosystem level CO2 395
17
17
uptake in the Arctic tundra, which is facilitated through structural changes in the canopy 396
(increased overall leaf area), while on a leaf level, the photosynthetic capacity remains 397
unchanged. 398
399
Foliar respiration 400
In contrast to photosynthetic parameters, 50% higher Narea values corresponded with 50% 401
higher Rd rates for the N+P treatment in B. nana while for E. vaginatum respiration was 27% 402
higher with a 15% increase in Narea or the N+P treatment (Fig. 3e). The lack of increased 403
respiration in the N-only treatment (compared with the N+P leaves) of site 1 could be 404
explained by a lower P availability in this treatment, which for example in tropical forest 405
reduces Rd (Meir et al. 2001), although we do not have foliar P data to confirm this. Increased 406
Rd after nutrient addition has been observed in other species as well (Manter et al. 2005), and 407
for B. nana this is a similar observation to Heskel et al. (2012). However, different from the 408
latter study, we only observed a significant increase in respiration after < 20 years of N and P 409
addition, and not after a shorter duration of the experiment. Heskel et al. (2012) also observed 410
increased numbers of mitochondrial area (density and size) in E. vaginatum and B. nana after 411
N+P addition. Since investments in mitochondria require more N, this could partially explain 412
why leaves with a higher Narea have higher Rd values. Additionally, if the excess foliar N is 413
invested in more non-mitochondrial proteins, this could cause higher maintenance respiration 414
due to higher protein turnover rates (Penning de Vries 1975). 415
Overall, the results for Vcmax, Jmax and Rd show that for the investigated species the 416
different gas exchange parameters cannot be scaled with foliar N in a similar way. One 417
implication of higher foliar respiration with no increase in C-uptake in the fertilised leaves is 418
that less photosyntate is available for the metabolism in other parts of the plant and 419
ecosystem. Measuring the effects of long and short-term nutrient addition on whole plant 420
18
18
respiration rates (or ecosystem respiration) was beyond the scope of this study. However, 421
long term nutrient addition in the Arctic increases the aboveground biomass more than the 422
belowground (Mack et al. , 2004; Sullivan et al. 2007; Gough et al. , 2012), which can result 423
in relatively less belowground autotrophic respiration than aboveground (on the premise that 424
the respiration of belowground tissue would remain the same). Furthermore, N+P fertilisation 425
and N deposition can reduce microbial respiration, especially in the rhizosphere in temperate 426
ecosystems, which is a caused by decreased excretion of root exudates and/or decreases in 427
fine microbial biomass (Phillips and Fahey, 2007; Janssens et al. , 2010; Jia et al. , 2010). It is 428
therefore plausible that the increases in foliar respiration because of higher foliar N are 429
accompanied by decreases in respiration of other ecosystem compartments. We did not 430
measure the respiration rates of the other plant parts so cannot confirm this, but we suggest 431
that future studies on the influence of nutrient supply on Arctic C-budgets and C-fluxes 432
should include gas exchange measurements of all different ecosystem compartments. 433
434
Conclusions 435
Comparing two sites of different durations in N and P addition showed that the PNUE 436
decreases in both B. nana and E. vaginatum with increased N availability, while Rd increased 437
after long-term (> 20 years) and high dosage N addition. This either shows that for these two 438
species photosynthesis is either already highly efficient on a lea level scale, or that they 439
become limited for other nutrients with increasing N and P availability. This should be taken 440
into account when scaling photosynthetic parameters with foliar N data (though is probably 441
of less importance when scaling productivity for the Arctic with only LAI). Additionally, the 442
different results for photosynthetic parameters and foliar respiration show that both 443
parameters cannot be scaled with nutrient concentrations in a similar way, urging for 444
modelling both processes separately. Finally, this study showed that short-term effects (1-4 445
19
19
years) of nutrient addition on eco-physiological parameters cannot by default be extrapolated 446
to a decadal time scale. This underlines the importance and value of long-term ecological 447
experiments when we investigate the effects of environmental change on ecological 448
processes. 449
450
Acknowledgements 451
This work was funded by NSF grants from the division of Environmental Biology (Arctic 452
LTER Project) and from the office of Polar Programs (Arctic Natural Sciences, Arctic 453
Systems Science). We would also like to thank the Toolik Lake Field Station and the Arctic 454
LTER project (NSF-DEB-1026843) for logistical support. 455
456
References 457
Azcón- Bieto J and Osmond CB (1983) Relationship between photosynthesis and respiration 458
- the effect of carbohydrate status on the rate of CO2 production by respiration in darkened 459
and illuminated wheat leaves. Plant Physiology 71: 574-581 460
Bauer GA, Bazzaz FA, Minocha R, Long S, Magill A, Aber J, Berntson GM (2004) Effects 461
of chronic N additions on tissue chemistry, photosynthetic capacity, and carbon sequestration 462
potential of a red pine (Pinus resinosa Ait.) stand in the NE United States. Forest Ecology 463
and Management 196 (1): 173–186 464
Bernacchi CJ, Singsaas EL, Pimentel C, Portis AR Jr and Long SP (2001) Improved 465
temperature response functions for models of Rubisco-limited photosynthesis. Plant, Cell and 466
Environment 24: 253–259 467
Bernacchi CJ, Portis AR, Nakano H, von Caemmerer S and Long SP (2002) Temperature 468
response of mesophyll conductance. Implications for the determination of Rubisco enzyme 469
20
20
kinetics and for limitations to photosynthesis in vivo. Plant Physiology 130: 1992–1998 470
Bernacchi CJ, Pimentel C and Long SP (2003) In vivo temperature response functions of 471
parameters required to model RuBP- limited photosynthesis. Plant, Cell and Environment 26: 472
1419– 1430 473
Bobbink R, Hicks K, Galloway J, Spranger T, Alkemade R, Ashmore M, Bustamante M, 474
Cinderby S, Davidson E, Dentener F, Emmett B, Erisman JW, Fenn M, Gilliam F, Nordin A, 475
Pardo L and De Vries W (2010) Global assessment of nitrogen deposition effects on 476
terrestrial plant diversity: a synthesis. Ecological Applications 20: 30-59 477
Boelman NT, Stieglitz M, Rueth HM, Sommerkorn M, Griffin KL, Shaver GR. and Gamon J. 478
A (2003) Response of NDVI, biomass, and ecosystem gas exchange to long-term warming 479
and fertilization in wet sedge tundra. Oecologia 135: 414-421 480
Bret-Harte MS, Shaver GR, Zoerner JP, Johnstone JF, Wagner JL, Chavez AS, Gunkelman 481
RF, Lippert SC, and Laundre JA (2001) Developmental plasticity allows Betula nana to 482
dominate tundra subjected to an altered environment. Ecology 82: 18-32 483
Bret-Harte MS, Shaver GR and Chapin FS (2002) Primary and secondary stem growth in 484
arctic shrubs: implications for community response to environmental change. Journal of 485
Ecology 90: 251-267 486
Britton ME (1966) Vegetation of the Arctic Tundra. In: Hanson, H. P. (ed.) Arctic Biology. 487
Oregon State University Press, pp. 67-130 488
Bubier JR, Smith R, Juutinen S, Moore T, Minocha R, Long S and Minocha (2011). Effects 489
of nutrient addition on leaf chemistry, morphology, and photosynthetic capacity of three bog 490
shrubs. Oecologia: 167:355–368 491
21
21
Chapin FS and Shaver GR (1996) Physiological and growth responses of arctic plants to a 492
field experiment simulating climatic change. Ecology 77: 822-840 493
Chapin FS, Shaver GR, Giblin AE, Nadelhoffer KJ and Laundre JA (1995) Responses of 494
arctic tundra to experimental and observed changes in climate. Ecology 76: 694-711 495
Ellsworth DS and Reich PB (1993) Canopy structure and vertical patterns of photosynthesis 496
and related leaf traits in a deciduous forest. Oecologia 96: 169-178 497
Evans JR (1989) Photosynthesis and nitrogen relationships in leaves of C-3 plants. Oecologia 498
78: 9-19 499
Evans JR and Poorter H (2001) Photosynthetic acclimation of plants to growth irradiance: the 500
relative importance of specific leaf area and nitrogen partitioning in maximizing carbon gain. 501
Plant Cell and Environment 24: 755-767 502
Farquhar GD, Von Caemmerer S and Berry JA (1980) biochemical model of photosynthetic 503
CO2 assimilation in leavesof C 3 species. Planta 149: 78-90 504
Field CB and Mooney HA (1986) The photosynthesis-nitrogen relationship in wild plants. In: 505
T.J, G. (ed.) On the economy of plant form and function. Cambridge University Press, pp. 506
25–55 507
Frey KE, McClelland JW, Holmes RM and Smith LC (2007) Impacts of climate warming and 508
permafrost thaw on the riverine transport of nitrogen and phosphorus to the Kara Sea. Journal 509
of Geophysical Research-Biogeosciences 112 510
Friend A, Geider R, Behrenfeld M and Still C (2009) Photosynthesis in Global-Scale Models. 511
In: Laisk, A., Nedbal, L. and Govindjee (eds.), Photosynthesis in silico. Springer 512
Netherlands, pp. 465-497 513
22
22
Gough L, Moore JC, Shaver GR, Simpson RT and Johnson DR (2012) Above- and 514
belowground responses of arctic tundra ecosystems to altered soil nutrients and mammalian 515
herbivory. Ecology 93: 1683-1694 516
Heskel MA, Anderson OR, Atkin OK, Turnbull MH and Griffin KL (2012) Leaf- and cell-517
level carbon cycling responses to a nitrogen and phosphorus gradient in two arctic tundra 518
species. American Journal of Botany 99: 1702-1714 519
Hobbie SE, Gough L, Shaver GR (2005) Species compositional differences on different-aged 520
glacial landscapes drive contrasting responses of tundra to nutrient addition. Journal of 521
Ecology 93: 770-782 522
Janssens IA, Dieleman W, Luyssaert S, Subke JA, Reichstein M, Ceulemans R, Ciais P, 523
Dolman AJ, Grace J, Matteucci G, Papale D, Piao SL, Schulze ED, Tang J and Law BE 524
(2010) Reduction of forest soil respiration in response to nitrogen deposition. - Nature Geosci 525
3: 315-322 526
Jia SX, Wang ZQ, Li XP, Sun Y, Zhang XP and Liang AZ (2010) N fertilization affects on 527
soil respiration, microbial biomass and root respiration in Larix gmelinii and Fraxinus 528
mandshurica plantations in China. Plant and Soil 333: 325-336 529
Johnson LC, Shaver GR, Cades DH, Rastetter E, Nadelhoffer K, Giblin A, Laundre J and 530
Stanley A (2000) Plant carbon-nutrient interactions control CO2 exchange in Alaskan wet 531
sedge tundra ecosystems. Ecology 81: 453-469 532
Kattge J, Knorr W, Raddatz T and Wirth C (2009) Quantifying photosynthetic capacity and 533
its relationship to leaf nitrogen content for global-scale terrestrial biosphere models. Global 534
Change Biology 15: 976-991 535
23
23
Keuper F, van Bodegom PM, Dorrepaal E, Weedon JT, van Hal J, van Logtestijn RSP and 536
Aerts R (2012) A frozen feast: thawing permafrost increases plant-available nitrogen in 537
subarctic peatlands. Global Change Biology 18: 1998-2007 538
Li Y, Ren B, Ding L, Shen Q, Peng S and Guo S (2013) Does chloroplast size influence 539
photosynthetic nitrogen use efficiency? PLoS One 8: e62036 540
Mack MC, Schuur EAG, Bret-Harte MS, Shaver GR and Chapin FS (2004) Ecosystem 541
carbon storage in arctic tundra reduced by long-term nutrient fertilization. Nature 431: 440-542
443 543
Manter DK, Kavanagh KL and Rose CL( 2005) Growth response of Douglas-fir seedlings to 544
nitrogen fertilization: importance of Rubisco activation state and respiration rates. Tree 545
Physiology 25: 1015–1021 546
Matthes-Sears U, Matthes-Sears WC, Hastings SJ and Oechel WC (1988) The effects of 547
topography and nutrient status on the biomass, vegetative characteristics, and gas exchange of 548
two deciduous shrubs on an arctic tundra slope. Arctic Antarctic and Alpine Research 20: 549
342-351 550
McClelland JW, Stieglitz M, Pan F, Holmes RM and Peterson BJ (2007) Recent changes in 551
nitrate and dissolved organic carbon export from the upper Kuparuk River, North Slope, 552
Alaska. Journal of Geophysical Research-Biogeosciences 112 553
24
24
Meir P, Grace J and Miranda AC (2001) Leaf respiration in two tropical rainforests: 554
constraints on physiology by phosphorus, nitrogen and temperature. Functional Ecology 555
15(3): 378-387 556
Meir P, Kruijt B, Broadmeadow M, Barbosa E, Kull O, Carswell F, Nobre A and Jarvi, PG 557
(2002) Acclimation of photosynthetic capacity to irradiance in tree canopies in relation to leaf 558
nitrogen concentration and leaf mass per unit area. Plant Cell and Environment 25: 343-357 559
Niinemets U (2003) Role of foliar nitrogen in light harvesting and shade tolerance of four 560
temperate deciduous woody species. Functional Ecology 11(4): 518–531 561
Niinemets U (2007) Photosynthesis and resource distribution through plant canopies. Plant 562
Cell and Environment 30: 1052-1071 563
Oberbauer SF, Hastings SJ, Beyers JL and Oechel WC (1989) Comparative effects of 564
downslope water and nutrient movement on plant nutrition, photosynthesis, and growth in 565
alaskan tundra. Holarctic Ecology 12: 324-334. 566
Penning De Vries FWT (1975) The Cost of Maintenance Processes in Plant Cells. Annals of 567
Botany 39: 77-92 568
Phillips, R. P. and Fahey, T. J. 2007. Fertilization effects on fineroot biomass, rhizosphere 569
microbes and respiratory fluxes in hardwood forest soils. - New Phytologist 176: 655-664 570
Porra RJ, Thompson WA and Kriedemann PE (1989) Determination of accurate extinction 571
coefficients and simultaneous-equations for assaying chlorophyll-a and chlorophyll-b 572
extracted with 4 different solvents - verification of the concentration of chlorophyll standards 573
by atomic-absorption spectroscopy. Biochimica Et Biophysica Acta 975: 384-394 574
25
25
Sharkey TD, Bernacchi CJ, Farquhar GD and Singsaas EL (2007) Fitting photosynthetic 575
carbon dioxide response curves for C-3 leaves. Plant Cell and Environment 30: 1035-1040 576
Shaver GR, Bret-Harte SM, Jones MH, Johnstone J, Gough L, Laundre J and Chapin FS 577
(2001) Species composition interacts with fertilizer to control long-term change in tundra 578
productivity. Ecology 82: 3163-3181 579
Shaver GR and Chapin FS (1980) Response to fertilization by various plant-growth forms in 580
an alaskan tundra - nutrient accumulation and growth. Ecology 61: 662-675 581
Shaver GR and Chapin FS (1986) Effect of fertilizer on production and biomass of tussock 582
tundra, Alaska, USA. Arctic and Alpine Research 18: 261-268 583
Shaver GR and Chapin FS (1991) Production - biomass relationships and element cycling in 584
contrasting arctic vegetation types. Ecological Monographs 61: 1-31 585
Shaver GR and Chapin FS (1995) Long-term responses to factorial, NPK fertilizer treatment 586
by alaskan wet and moist tundra sedge species. Ecography 18: 259-275 587
Shaver GR, Johnson LC, Cades DH, Murray G, Laundre JA, Rastetter EB, Nadelhoffer KJ 588
and Giblin AE (1998) Biomass and CO2 flux in wet sedge tundras: Responses to nutrients, 589
temperature, and light. Ecological Monographs 68: 75-97 590
Shaver GR, Rastetter EB, Salmon V, Street LE, van de Weg MJ, Rocha A, van Wijk MT 591
(2013) Pan-Arctic modelling of net ecosystem exchange of CO2. Philosophical Transactions 592
of the Royal Society B. (in press). 593
Sims DA and Gamon JA (2002) Relationships between leaf pigment content and spectral 594
reflectance across a wide range of species, leaf structures and developmental stages. Remote 595
Sensing of Environment 81: 337-354 596
26
26
Street LE, Shaver GR, Williams M and Van Wijk MT (2007) What is the relationship 597
between changes in canopy leaf area and changes in photosynthetic CO2 flux in arctic 598
ecosystems? Journal of Ecology 95: 139-150 599
Thornton PE, Lamarque JF, Rosenbloom NA and Mahowald NM (2007) Influence of carbon-600
nitrogen cycle coupling on land model response to CO2 fertilization and climate variability. 601
Global Biogeochemical Cycles 21: GB4018 602
Van Heerwaarden LM, Toet S and Aerts R (2003) Nitrogen and phosphorus resorption 603
efficiency and proficiency in six sub-arctic bog species after 4 years of nitrogen fertilization. 604
Journal of Ecology 91: 1060-1070 605
Van Wijk MT, Williams M and Shaver GR (2005) Tight coupling between leaf area index 606
and foliage N content in arctic plant communities. Oecologia 142: 421–427 607
von Caemmerer S (2000) Biochemical models of leaf photosynthesis. CSIRO Publishing 608
Wullschleger SD (1993) Biochemical limitations to carbon assimilation in C3 Plants-A 609
retrospective analysis of the A/Ci curves from 109 species. Journal o Experimental Botany 610
44: 907-920 611
Zaehle S and Dalmonech D (2011) Carbon–nitrogen interactions on land at global scales: 612
current understanding in modelling climate biosphere feedbacks. Current Opinion in 613
Environmental Sustainability 3: 311-320 614
Zaehle S and Friend AD (2010) Carbon and nitrogen cycle dynamics in the O-CN land 615
surface model: 1. Model description, site-scale evaluation, and sensitivity to parameter 616
estimates. Global Biogeochem. Cycles 24: GB1005 617
27
27
Zamin TJ and Grogan P (2012) Birch shrub growth in the low Arctic: the relative importance 618
of experimental warming, enhanced nutrient availability, snow depth and caribou exclusion. - 619
Environmental Research Letters 7 620
621
28
28
Table 1. Overview of the nutrient addition treatments and their codes from the two different 622
sites used in this study. 623
Plot code Treatment Annual N addition
(g m-2 yr-1)
Annual P
addition
(g m-2 yr-1)
Duration of
nutrient
addition
Site 1
CT Control 0 0 0 years
NP N + P addition 10 5 22 years
N N addition 10 0 22 years
P P addition 0 5 22 years
Site 2
CT Control 0 0 0 years
F10 N + P addition 10 5 5 years
F05 N + P addition 5 2.5 5 years
YR1 N + P addition 10 5 6 weeks (first
year)
624
625
29
29
Figure legends 626
Fig. 1 Diurnal average relative received PAR at three different canopy positions in the CT 627
and N+P plot of Site 1 throughout the day from 10 -30 July 2011 (left panes) (± standard 628
error, n=21), and the average received PAR per day (right panes). 629
630
Fig. 2 Foliar N on a mass and area basis per species and per site ± standard error (n=8). 631
Asterisks indicate a significant difference of a treatment from the control for that site and 632
species ( * = P<0.05, ** = P<0.01, ***= P<0.001). Abbreviations of the treatments are as in 633
Table 1. (CT= control, P = phosphorus addition only, N = nitrogen addition only, NP = 634
phosphorus and nitrogen addition, YR1 = first year of nutrient addition, F10 = 10 g N m-2 yr-635
1, F05= 5 g N m-2yr-1) 636
637
Fig. 3 Foliar CO2 exchange parameters (Vcmax, J max and Rd) and nitrogen content (Narea) after 638
long term nutrient addition (Site 1) or short term nutrient addition (Site 2) for E. vaginatum 639
(closed circles) and B. nana (open circles) ± standard error (n=8). Abbreviations of the 640
treatments are as in Table 1. Asterisks indicate that for that treatment the Y-axis parameters is 641
significantly different (P<0.05, Dunnet’s post-hoc test) from the CT treatment of that site and 642
species. Open circles represent B. nana and closed circles E. vaginatum. 643
644
Fig. 4 Foliar chlorophyll (a+b) and leaf mass per area (LMA) and nitrogen content (Narea) 645
after long term nutrient addition (Site 1) or short term nutrient addition (Site 2) for E. 646
vaginatum (closed circles) and B. nana (open circles) ± standard error (n=8). Abbreviations 647
of the treatments are as in Table 1 and Fig. 2. Asterisks indicate that for that treatment the Y-648
axis parameters is significantly different (P<0.05, Dunnet’s post-hoc test) from the CT 649
30
30
treatment of that site and species. Open circles represent B. nana and closed circles E. 650
vaginatum. 651
652
Fig. 5 Foliar N on a mass (a) and area basis (b), CO2 exchange parameters (Vcmax, Jmax ) on an 653
area basis (c and d), chlorophyll content (e), and leaf mass per area (LMA, f) for B. nana in 654
the NP treatment of Site 1 (>20 years of N+P addition, open circles and dashed line for trend 655
line) and the CT treatment (closed circles, solid line as the trend line). X-axes represent the 656
standardized level of photosynthetic active radiation (PAR) received (1=top canopy). 657
658
31
31
Figure 1 659
660
661
32
32
Figure 2 662
663
664
665
33
33
Figure 3 666
667
668
669
34
34
670 Figure 4 671
672
673
674
35
35
Figure 675
5676
677